Introduction

Allogeneic haematopoietic stem cell transplantation (SCT) remains the most effective curative option for a variety of haematological disorders such as leukaemias and bone marrow failure syndromes. Owing to the discrepancy between demand and supply of human leucocyte antigen (HLA) matched stem cell donors, it is necessary in many cases to transplant stem cells from HLA mismatched donors.^{1, 2, 3, 4} However, this results in an increased incidence of life-threatening graft-versus-host disease (GVHD), graft failure and severe viral infections.^{5, 6} Recently, we showed that there is an upper limit to the degree of major histocompatibility complex (MHC) class I amino-acid sequence disparity that elicits an allogeneic CTL response in vitro.⁷ In a cytotoxic T-lymphocyte precursor (CTLp) assay, HLA-C differences with five or more amino-acid differences in the -helices as well as five or more in the -sheet (55) did not elicit an allogeneic CTL response in all but one case. This finding might be useful when selecting haematopoietic stem cell donors for transplantation, as several studies have shown the CTLp assay to be a clinically relevant parameter for the assessment of SCT outcome.^{8, 9, 10, 11, 12, 13}

An upper limit to the degree of MHC amino-acid sequence disparity recognizable by T cells is in agreement with the theories on T-cell selection processes within the thymus. In generating a T-cell repertoire with a sufficiently narrow responsiveness for self-MHC, positive thymic selection limits the capacity to recognize MHC molecules the structure and sequence of which have diverged extensively.^{14, 15, 16, 17, 18} Many 55 mismatched MHC class I molecules fall into this category of extensively diverged MHC molecules. If no T-cell alloreactivity against a 55 mismatch can be measured in vitro, we expect that transplantation across a 55 mismatch will not lead to T-cell alloreactivity after transplantation in vivo.

We therefore questioned whether there is a difference between the prognostic value of a negative CTLp assay performed on a single 55 MHC class I mismatched pair and a single less-diverged MHC class I mismatched pair. Not all peptides presented by MHC class I in the recipient have been presented by the recipient's peripheral blood lymphocytes in the CTLp assay.¹⁹ If less-diverged class I MHC on other cell types in the recipient presents peptides specific for these tissues, an in vivo CTL alloimmune response could be triggered, even in the case of a negative CTLp assay before transplantation, as this test is performed on lymphocytes. Whether this is the case for 55 mismatched MHC is open to question. If the donor T-cell repertoire is unresponsive to recipient 55 mismatched MHC, then it should not matter which peptide is presented by the 55 MHC of the recipient.

To investigate whether SCT with a single 55 MHC class I mismatched graft leads to successful transplant outcome, we analysed the follow-up records of a group of Dutch patients transplanted with a single MHC class I mismatched graft. As the 55 was not known at the time of transplantation, there is no selection bias. In this report we address two questions. The first is whether we can subdivide the donor–recipient pairs with a negative CTLp assay into a prognostic favourable and unfavourable group by distinction between the pairs being respectively 55 MHC class I mismatched or not. Secondly, can the 55 MHC class I mismatch category replace the CTLp assay as a prognostic tool for SCT outcome.

Materials and methods

Donor–recipient pairs

The studied cohort of donors and recipients that were transplanted between 1992 and January 2004 contained 74 pairs. The transplants, registered by the Europdonor Foundation, took place in the following Dutch haematopoietic SCT centres: Leiden University Medical Centre in Leiden (n=40), Erasmus MC/Daniel den Hoed in Rotterdam (n=23) and University Medical Centre, Wilhelmina Children's Hospital, in Utrecht (n=11).

The donors originated from the national donor registry of the Europdonor Foundation, International donor registries or were related individuals. The follow-up data were obtained from the Dutch national transplantation registry Typhon. The median follow-up of all patients was 1 year (0.1–10 years). The median follow-up of the surviving patients was 4 years (1–10 years).

In our centre, the CTLp assay is routinely used as a tool to select the most suitable stem cell donor for a patient. However, we had to exclude 21 couples for whom the CTLp assay could not be performed owing to an insufficient number of available donor or recipient peripheral blood lymphocytes (PBL) (n=8) or because the test failed (n=13). The patient's PBL do not always meet the requirements for the CTLp assay because of the underlying disease and some therapies to treat the disease affect the quality of PBL. The donor PBL are sometimes transported over long distances resulting in suboptimal quality of the cells.

Diagnosis at the time of transplantation, age of recipients and donors, cytomegalovirus (CMV) serology, preconditioning and GVHD prophylaxis of the 53 included pairs are shown in Table 1a and b. The following diagnoses were categorized in this study as conditions with a high risk for transplant-related mortality: acute lymphocytic leukaemia (ALL) or acute myelogenous leukaemia (AML) beyond first remission or in relapse; chronic myelogenous leukaemia (CML) in second chronic phase, accelerated phase or in blast phase; severe aplastic anaemia (SAA), Fanconi anaemia (FA) and myelodysplastic syndrome (MDS) in case of life-threatening haemorrhage and/or refractoriness to platelet transfusion or infection. The other diagnoses were considered low risk for transplant-related mortality.

Table 1 - (a) Patient and donor characteristics and (b) treatment.

Full table

Human leucocyte antigen typing and donors–recipient matching

The donors and recipients were typed at high resolution for the loci HLA-A, -B, -C, -DRB1, -DQB1 and -DPB1 as described before.²⁰ The following techniques were used: the polymerase chain reaction sequence specific primer (PCR-SSP) for high-resolution allele typing and sequence-based typing (SBT) for part of the HLA-C alleles. All pairs were chosen because they had a single HLA-A, -B or -C antigen/allele mismatch in the graft-versus-host direction; 19 were mismatched for HLA-A, 6 for HLA-B and 28 for HLA-C (Table 2). All pairs were matched for the HLA-DRB1 and-DQB1 alleles. As HLA-A, -B and -C mismatches have a similar adverse effect in SCT,⁶ we did not separate these groups in the analysis. Additional matching for HLA-DPB1 was not possible, owing to the small number of pairs. HLA-DPB1 mismatches were not included in the analysis as previous studies showed that they were hardly relevant to CTLp assay outcome.⁷

Table 2 - Number and type of mismatched MHC class I loci.

Full table

Amino-acid sequencing of the mismatched MHC class I molecules

The amino-acid sequences were obtained from the website of the European Bioinformatics Institute (http://www.ebi.ac.uk/imgt/hla/). The mismatched MHC molecules were examined for amino-acid differences (substitutions) on the 1 and 2 domain (positions 1–182). The 55 MHC class I mismatches were defined as having at least five amino-acid differences in the -helices plus at least five in the -sheet. The -helices of MHC class I consist of positions 50–85 and 138–179 and the -sheet of MHC class I is determined by positions 4–12, 21–28, 32–37, 94–102, 112–118 and 123–126. This definition is based on our previous work where we observed that single MHC class I mismatches falling into the 55 category were associated with an undetectable T-cell alloreactivity in the CTLp assay.⁷

Cytotoxic T-lymphocyte precursor assay

The CTLp assay was performed as described by Zhang et al.²¹ and Oudshoorn et al.²⁰ The target cells were labelled with ⁵¹Cr. The reproducibility of this assay has been described previously.²¹ Each assay was performed in the graft-versus-host (GVH) direction, that is, donor cells were used as the responder and the recipient cells as the stimulator and target. Cultures showing higher lysis than three standard deviations above the mean of the spontaneous release (target cells without any responder cells) were considered to be positive. The frequency of CTLp and 95% confidence interval were calculated as described by Taswell^{22, 23} and Strijbosch et al.²⁴ Only those experiments that showed P>0.05 using the jack-knife method were accepted. Thus the relationship between responder cell dose and the number of non-responding wells was considered to be consistent with a single-hit kinetic model. The CTLp assay was defined negative if the CTLp frequency was 1 per 10⁶ PBL and positive if the CTLp frequency was >1 per 10⁶ PBL. In a previous analysis, we considered a CTLp frequency of >5 per 10⁶ PBL as the cutoff point for high response because in that study we aimed at a clear separation between the positive and negative group by not considering CTL between 1 and 4.⁷ This study focuses on the group with CTLp 1 per 10⁶ PBL compared to other transplants with CTLp >1 per 10⁶ PBL.

Conditioning regimen and transplantation

Conditioning therapy before SCT consisted of total body irradiation (TBI) in combination with cyclophosphamide in 34 cases, or a combination of busulphan and cyclophosphamide in 15 cases. Four patients received non-myeloablative conditioning therapy, either TBI and fludarabine (n=1) or BEAM (n=3). Cyclosporine was used as GVHD prophylaxis in 18 cases, and complemented with in vivo T-cell depletion in 15 cases and methotrexate in 18 cases. Two patients received in vivo T-cell depletion. The recipients were given bone marrow–derived stem cells in 42 cases and peripheral blood harvest after G-CSF in 11 cases. Grafts were T-cell depleted with Campath (n=18), CD34+ selection (n=10) or other methods (n=16). Nine patients received non-T-cell-depleted grafts.

Statistics

All statistical analyses were performed using SPSS 10.0. The ² tests (for categorical variables) and Mann–Whitney U test (for continuous variables) were used to compare recipient-, disease- and transplantation-related variables between the groups with a positive and negative CTLp assay outcome. Overall survival after transplantation was analysed with Kaplan–Meier curves (surviving patients were censored at last contact or at the time of second transplantation). Univariate and multivariate analyses were performed with the Cox proportional hazards model. Other end points were the following: (1) transplant-related mortality defined as time of death without evidence of disease recurrence; (2) disease relapse defined as disease recurrence; and (3) incidence of acute (grade II–IV) and chronic GVHD.^{25, 26, 27}

Results

The patients

Overall survival in this cohort of 53 donor–recipient pairs was 44% after 4 years. HLA-A and HLA-C mismatches were well represented in this group with respectively 19 and 28 pairs, while only 6 pairs were transplanted over an HLA-B mismatch, of which all had few amino-acid differences (Table 2). Fifteen pairs in this cohort were 55 mismatched. The 55 mismatched group was not predominated by HLA-C mismatches. The CTLp assay in the GVH direction was negative for 31 pairs and positive for 22 pairs. Of the 15 55 mismatched pairs, 12 had a negative CTLp assay outcome. Between the CTLp-positive and -negative group, donor–recipient-related parameters such as gender, gender match, age and diagnosis were equally distributed, and there was no significant difference in patient conditioning and transplantation regimen parameters (Table 1a and b).

CTLp assay, single 55 MHC class I mismatches and survival

We observed a strong correlation between negative CTLp frequencies and overall patient survival after SCT in the single MHC class I mismatched setting (Figure 1). Four-year survival was 63% in the case of negative CTLp frequencies and 20% in the case of positive CTLp frequencies (hazard ratio=2.57; 95% CI=1.206–5.478; P=0.014).

Figure 1.

Overall patient survival after SCT correlated with the CTLp assay outcome. The number of pairs in each group: 31 pairs with a negative CTLp assay and 22 pairs with a positive CTLp assay. The overall survival of patients correlated with the CTLp assay outcome. Positive CTLp frequencies had a hazard ratio of 2.57 (95% CI=1.206–5.478; P=0.014) compared to a negative CTLp assay.

Full figure and legend (11K)

Within the CTLp-negative group, donor–recipient pairs with a 55 mismatch demonstrated superior overall survival (Figure 2). Four-year survival was 80% after SCT with a 55 mismatched graft and negative CTLp frequencies (hazard ratio=0.144; 95% CI=0.033–0.633; P=0.010), and 47% in the case of other MHC class I differences plus negative CTLp frequencies. The occurrence of acute GVHD, chronic GVHD and disease relapse and causes of patient mortality are shown in Table 3. Ninety per cent of transplant-related mortality was caused by infections, probably owing to a compromised immune system. We found no statistically significant relation between the CTLp assay outcome or mismatches and these parameters (Table 4).

Figure 2.

Overall patient survival after SCT correlated with the CTLp assay outcome and 55 mismatch category. The number of pairs in each group: 12 pairs with a 55 MHC class I difference and negative CTLp assay, 19 pairs with another single MHC class difference and a negative CTLp assay and 22 pairs with a positive CTLp assay. Recipients with a negative CTLp assay outcome and a 55 MHC class I difference with the donor had a superior chance of survival compared to the other groups (hazard ratio=0.144; 95% CI=(0.033–0.633); P=0.010).

Full figure and legend (13K)

Table 3 - SCT outcomes.

Full table

Table 4 - Univariate analysis on SCT outcomes.

Full table

In multivariate analysis, the benefit for recipients of a 55 mismatched graft in combination with a negative CTLp assay was more apparent (Table 5). Multivariate analysis revealed a second important parameter. Transplanting a single MHC class I mismatched graft from a female donor to a male recipient had significant adverse effect on transplantation outcome independent of CTLp frequencies and MHC class I mismatch categories. The mortality rate of this group was 11 out of 13 patients compared to 17 out of the 40 in the other group. The other parameters in the multivariate analysis – risk status of the diagnosis, CMV infection, age of the recipient and T-cell depletion of the graft – did not seem to have a significant additional impact on overall survival (Table 5).

Table 5 - Results from multivariate analysis on patient survival after SCT.

Full table

MHC class I mismatch categories and survival

Irrespective of the CTLp assay outcome, 4-year overall survival of recipients of a 55 mismatched graft was 65% compared to 37% of recipients of another MHC class I mismatched graft (Figure 3). This was not statistically significant (hazard ratio=0.440; 95% CI=0.167–1.160; P=0.1). In multivariate analyses, however, there was a significant correlation. The 55 mismatch had a hazard ratio of 0.316 (95% CI=0.16–0.81; P=0.014) on overall patient survival. Transplanting a graft from a female donor to a male recipient had a hazard ratio of 3.408 (95% CI: 1.53–7.68; P=0.003), while risk status of the diagnosis had a hazard ratio of 1.150 (95% CI: 0.49–2.70; P=0.8), CMV infection had a hazard ratio of 1.343 (95% CI=0.60–2.99; P=0.5), age of the recipient had a hazard ratio of 1.002 (95% CI: 0.98–1.03; P=0.9) and T-cell depletion of the graft had a hazard ratio of 1.361 (95% CI: 0.54–3.45; P=0.5). The CTLp frequencies were not included in this multivariate analysis because the category 55 and CTLp frequencies are related parameters.⁷

Figure 3.

Overall survival of patients related to 55 MHC class I mismatch. The number of pairs in each group: 15 pairs with a 55 MHC class I difference, 38 pairs with another single MHC class difference. The results were not statistically significant (hazard ratio=0.440; 95% CI=0.167–1.160; P=0.1).

Full figure and legend (11K)

Overall survival of the 21 pairs excluded from this study was 40% after 4 years, which is similar to the pairs included within this study. Nine pairs had a 55 mismatch. There was a trend towards better survival for 55 mismatched pairs (data not shown).

Discussion

In the present report, we demonstrate the prognostic value of the CTLp assay for the outcome of transplantation with single MHC class I mismatched grafts. In agreement with earlier studies, positive CTLp frequencies are associated with a strong adverse effect on SCT outcome in comparison to negative CTLp frequencies.^{8, 9, 10, 11, 12, 13} A significant number of transplants with negative CTLp frequencies, however, also led to poor outcome. Combining the MHC sequence difference categories with the in vitro CTLp results proved to be more informative and appears to enhance the predictability of SCT outcome. A single 55 MHC class I mismatch in addition to a negative CTLp frequency was associated with superior survival after SCT, compared to other single MHC differences. No significant effect on the occurrence of GVHD and relapse was observed, which may be explained by the application of T cell depletion as well as the relatively low incidence of these end points.

Comparison between the negative CTLp assay and 55 category shows that the prognostic value of the 55 MHC class I mismatch is similar to that of the CTLp assay. Four-year overall survival was 63% in the case of a negative CTLp assay and 65% in the case of a 55 mismatched graft, which is not significant. In addition, survival was 20% for CTLp-positive patients and 37% for patients with another mismatch. The 55 category, however, is less laborious than the CTLp assay. However, considering our results, we are of the opinion that additional in vitro testing is important when using the 55 category.

There has been considerable controversy regarding the relationship between CTLp and transplant outcomes. The best correlations have been observed after T-cell depletion, because in that situation only high responders can cause T-cell alloreactivity after T-cell depletion.^{8, 9, 10, 11, 12, 13} Without T-cell depletion, low responders are also able to induce T-cell alloreactivity. Most grafts in this study were T-cell depleted.

Our findings seem to be in contradiction with the widely propagated idea that MHC mismatches with few amino-acid sequence differences are more advantageous in SCT than mismatches with more sequence differences.^{5, 6, 28} An upper limit to the degree of MHC sequence disparity that is still able to elicit an allogeneic CTL response does not, however, exclude a lower limit of allorecognition. Secondly, not all mismatches with many amino-acid sequence differences fall into the 55 category, as not all mismatches have many differences on both the -helices and -sheet.⁷ Such mismatches do lead to T-cell alloreactivity. Thirdly, some amino-acid positions have been described as important for T-cell alloreactivity, such as 116.^{20, 29} Some 55 mismatches have a 116 difference and some do not. Previously, we have demonstrated that mismatches at position 116 do not correlate with T-cell alloreactivity in vitro⁷ and in this study they do not affect survival in a different way than other mismatches. A problem that could arise in some cases is that a 55 mismatched MHC class I molecule could lead to deficient recognition of pathogenic antigens, which is less likely to occur in cases with few amino-acid sequence differences. However, in the case of single class I mismatched donor–recipient pairs, the matched class I molecules can probably fulfil this task.

Although alloreactive CTL are considered to be responsible for SCT-related clinical complications, NK cells are also able to lyse allogeneic cells, if the target cells lack the inhibitory killer immunoglobulin-like receptor (KIR) ligand motifs that the responder cells have.^{30, 31} Highly diverged MHC class I mismatches as 55 are more likely to be KIR ligand mismatched and thus could elicit NK responses. A combination of an HLA-Bw6 recipient and an HLA-Bw4 donor is considered as an inhibitory KIR ligand mismatched pair. In case of an HLA-C mismatch, the absence or presence of the Ser77 and Asn80 motif (C1; ligand for the inhibitory KIR2DL2 and KIR2DL3) and Asn77 and Lys80 motif (C2; ligand for the inhibitory KIR2DL1) was compared between donor and recipient.^{32, 33} In this setting, we did not find a relation between inhibitory KIR ligand mismatches and survival (data not shown). Owing to the small number of inhibitory KIR ligand mismatched pairs in this study (n=5), we can only conclude that it did not interfere with our results. The role of alloreactive NK cells in SCT remains controversial, as it has been described to be both unfavourable and beneficial for successful SCT outcome.^{34, 35, 36, 37, 38, 39, 40}

In our patient cohort, a 55 effect had a greater impact on recipient survival than most of the other factors such as age of the recipient, high risk status of the haematological disorder, CMV infection and T-cell depletion of the graft, which all have been described to influence successful SCT outcome.^{41, 42, 43} We could confirm the adverse effect of the female donor–male recipient combination.⁴⁴ Selection of such a single MHC class I mismatched female donor instead of selecting a male donor with a similar HLA mismatch cannot be recommended. This parameter affecting SCT outcome did not correlate with in vitro CTL alloreactivity before transplantation. The absence of such an association may be explained by the hypothesis that not all peptides presented by MHC class I in the recipient are also presented by the peripheral blood lymphocytes in the CTLp assay. Another possibility is that it depends on other cell types of the immune system. However, many studies have described the significance of the cytotoxic T-cell response against male minor histocompatibility antigens presented by either matched or mismatched recipient MHC.^{45, 46, 47, 48}

We did not include all clinical parameters in the multivariate analysis, owing to the small number of patients studied. Some parameters were homogeneous, like the conditioning as only a few patients received non-myeloablative treatment. Most patients received T-cell-depleted grafts. However, many different T-cell depletion methods were used. Comparison between these methods was not possible. A large multi-centre and multivariate analysis of SCT following the criteria presented here will elucidate the threshold of sequence differences that do not elicit T-cell alloreactivity. It might be suitable to include structural data on MHC molecules and functional similarity of amino acids.

In conclusion, the present analysis demonstrates that there is an upper threshold to the sequence differences of MHC class I mismatches that lead to T-cell alloreactivity and poor SCT outcome. Provided that the graft is single MHC class I mismatched, a highly diverged MHC class I mismatch can be permissible for transplantation and thus lead to successful SCT outcome; information obtained from in vitro analysis, however, remains imperative. This option of acceptable mismatches for SCT generates a larger pool of potentially compatible stem cell donors for patients lacking a fully MHC matched donor.

References

1. Dini G, Lanino E, Lamparelli T, Barbanti M, Sacchi N, Carcassi C et al. Unrelated donor marrow transplantation: initial experience of the Italian bone marrow transplant group (GITMO). Bone Marrow Transplant 1996; 17: 55–62. | PubMed | ChemPort |
2. O'Shea J, Cleaver S, Little AM, Madrigal A. Searching for an unrelated haemopoietic stem cell donor – a United Kingdom perspective. Clin Transplant 1999, 129–137.
3. Barker JN, Krepski TP, DeFor TE, Davies SM, Wagner JE, Weisdorf DJ. Searching for unrelated donor hematopoietic stem cells: availability and speed of umbilical cord blood versus bone marrow. Biol Blood Marrow Transplant 2002; 8: 257–260. | Article | PubMed | ISI |
4. Heemskerk MB, van Walraven SM, Cornelissen JJ, Barge RM, Bredius RG, Egeler RM et al. How to improve the search for an unrelated haematopoietic stem cell donor. Faster is better than more! Bone Marrow Transplant 2005; 35: 645–652. | ChemPort |
5. Petersdorf EW. HLA matching in allogeneic stem cell transplantation. Curr Opin Hematol 2004; 11: 386–391. | Article | PubMed | ChemPort |
6. Flomenberg N, Baxter-Lowe LA, Confer D, Fernandez-Vina M, Filipovich A, Horowitz M et al. Impact of HLA class I and class II high-resolution matching on outcomes of unrelated donor bone marrow transplantation: HLA-C mismatching is associated with a strong adverse effect on transplantation outcome. Blood 2004; 104: 1923–1930. | Article | PubMed | ISI | ChemPort |
7. Heemskerk MB, Roelen DL, Dankers MK, van Rood JJ, Claas FH, Doxiadis II et al. Allogeneic MHC class I molecules with numerous sequence differences do not elicit a CTL response. Hum Immunol 2005; 66: 969–976. | Article | PubMed | ChemPort |
8. Sharrock CE, Kaminski E, Man S. Limiting dilution analysis of human T cells: a useful clinical tool. Immunol Today 1990; 11: 281–286. | Article | PubMed | ISI | ChemPort |
9. Spencer A, Brookes PA, Kaminski E, Hows JM, Szydlo RM, van Rhee F et al. Cytotoxic T lymphocyte precursor frequency analyses in bone marrow transplantation with volunteer unrelated donors. Value in donor selection. Transplantation 1995; 59: 1302–1308. | PubMed | ISI | ChemPort |
10. Speiser DE, Loliger CC, Siren MK, Jeannet M. Pretransplant cytotoxic donor T-cell activity specific to patient HLA class I antigens correlating with mortality after unrelated BMT. Br J Haematol 1996; 93: 935–939. | Article | PubMed | ChemPort |
11. Keever-Taylor CA, Passweg J, Kawanishi Y, Casper J, Flomenberg N, Baxter-Lowe LA. Association of donor-derived host-reactive cytolytic and helper T cells with outcome following alternative donor T cell-depleted bone marrow transplantation. Bone Marrow Transplant 1997; 19: 1001–1009. | Article | PubMed | ChemPort |
12. van der MA, Joosten I, Schattenberg AV, de Witte TJ, Allebes WA. Cytotoxic T-lymphocyte precursor frequency (CTLp-f) as a tool for distinguishing permissible from non-permissible class I mismatches in T-cell-depleted allogeneic bone marrow transplantation. Br J Haematol 2000; 111: 685–694. | Article | PubMed |
13. Dolezalova L, Vrana M, Dobrovolna M, Loudova M, Cukrova V, Vitek A et al. Cytotoxic T lymphocyte precursor frequency analysis in the selection of HLA matched unrelated donors for hematopoietic stem cell transplantation: the correlation of CTLp frequency with HLA class I genotyping and aGVHD development. Neoplasma 2002; 49: 26–32. | PubMed | ISI | ChemPort |
14. Davis MM, Bjorkman PJ. T-cell antigen receptor genes and T-cell recognition. Nature 1988; 334: 395–402. | Article | PubMed | ISI | ChemPort |
15. Alam SM, Travers PJ, Wung JL, Nasholds W, Redpath S, Jameson SC et al. T-cell-receptor affinity and thymocyte positive selection. Nature 1996; 381: 616–620. | Article | PubMed | ISI | ChemPort |
16. Zerrahn J, Held W, Raulet DH. The MHC reactivity of the T cell repertoire prior to positive and negative selection. Cell 1997; 88: 627–636. | Article | PubMed | ISI | ChemPort |
17. Goldrath AW, Bevan MJ. Selecting and maintaining a diverse T-cell repertoire. Nature 1999; 402: 255–262. | Article | PubMed | ISI | ChemPort |
18. Sherman LA, Morgan DJ, Nugent CT, Hernandez FJ, Kreuwel HT, Murtaza A et al. Self-tolerance and the composition of T cell repertoire. Immunol Res 2000; 21: 305–313. | Article | PubMed | ISI | ChemPort |
19. de Bueger M, Bakker A, van Rood JJ, Van der WF, Goulmy E. Tissue distribution of human minor histocompatibility antigens. Ubiquitous versus restricted tissue distribution indicates heterogeneity among human cytotoxic T lymphocyte-defined non-MHC antigens. J Immunol 1992; 149: 1788–1794. | PubMed | ChemPort |
20. Oudshoorn M, Doxiadis II, van den Berg-Loonen PM, Voorter CE, Verduyn W, Claas FH. Functional versus structural matching: can the CTLp test be replaced by HLA allele typing? Hum Immunol 2002; 63: 176–184. | Article | PubMed | ISI | ChemPort |
21. Zhang L, Li SG, Vandekerckhove B, Termijtelen A, van Rood JJ, Claas FH. Analysis of cytotoxic T cell precursor frequencies directed against individual HLA-A and -B alloantigens. J Immunol Methods 1989; 121: 39–45. | Article | PubMed | ISI | ChemPort |
22. Taswell C. Limiting dilution assays for the determination of immunocompetent cell frequencies. I. Data analysis. J Immunol 1981; 126: 1614–1619. | PubMed | ISI | ChemPort |
23. Taswell C. Limiting dilution assays for the determination of immunocompetent cell frequencies. III. Validity tests for the single-hit Poisson model. J Immunol Methods 1984; 72: 29–40. | Article | PubMed | ISI | ChemPort |
24. Strijbosch LW, Buurman WA, Does RJ, Zinken PH, Groenewegen G. Limiting dilution assays. Experimental design and statistical analysis. J Immunol Methods 1987; 97: 133–140. | Article | PubMed | ISI | ChemPort |
25. Shulman HM, Sullivan KM, Weiden PL, McDonald GB, Striker GE, Sale GE et al. Chronic graft-versus-host syndrome in man. A long-term clinicopathologic study of 20 Seattle patients. Am J Med 1980; 69: 204–217. | Article | PubMed | ISI | ChemPort |
26. Sullivan M. Graft-versus-host-disease. In: Thomas ED, Blume KG, Forman SJ (eds). Bone Marrow Transplantation. Blackwell Scientific Publications: Cambridge, USA, 1994, pp 339–362.
27. Przepiorka D, Weisdorf D, Martin P, Klingemann HG, Beatty P, Hows J et al. 1994 consensus conference on acute GVHD grading. Bone Marrow Transplant 1995; 15: 825–828. | PubMed | ISI | ChemPort |
28. Petersdorf EW, Hansen JA, Martin PJ, Woolfrey A, Malkki M, Gooley T et al. Major-histocompatibility-complex class I alleles and antigens in hematopoietic-cell transplantation. N Engl J Med 2001; 345: 1794–1800. | Article | PubMed | ISI | ChemPort |
29. Ferrara GB, Bacigalupo A, Lamparelli T, Lanino E, Delfino L, Morabito A et al. Bone marrow transplantation from unrelated donors: the impact of mismatches with substitutions at position 116 of the human leukocyte antigen class I heavy chain. Blood 2001; 98: 3150–3155. | Article | PubMed | ISI | ChemPort |
30. Moretta A, Vitale M, Bottino C, Orengo AM, Morelli L, Augugliaro R et al. P58 molecules as putative receptors for major histocompatibility complex (MHC) class I molecules in human natural killer (NK) cells. Anti-p58 antibodies reconstitute lysis of MHC class I-protected cells in NK clones displaying different specificities. J Exp Med 1993; 178: 597–604. | Article | PubMed | ISI | ChemPort |
31. van der Meer A, Allebes WA, Paardekooper J, Ruiter J, Joosten I. HLA-C mismatches induce strong cytotoxic T-cell reactivity in the presence of an additional DRB/DQB mismatch and affect NK cell-mediated alloreactivity. Transplantation 2001; 72: 923–929. | Article | PubMed | ChemPort |
32. Colonna M, Borsellino G, Falco M, Ferrara GB, Strominger JL. HLA-C is the inhibitory ligand that determines dominant resistance to lysis by NK1- and NK2-specific natural killer cells. Proc Natl Acad Sci USA 1993; 90: 12000–12004. | Article | PubMed | ChemPort |
33. Mandelboim O, Reyburn HT, Vales-Gomez M, Pazmany L, Colonna M, Borsellino G et al. Protection from lysis by natural killer cells of group 1 and 2 specificity is mediated by residue 80 in human histocompatibility leukocyte antigen C alleles and also occurs with empty major histocompatibility complex molecules. J Exp Med 1996; 184: 913–922. | Article | PubMed | ISI | ChemPort |
34. Davies SM, Ruggieri L, DeFor T, Wagner JE, Weisdorf DJ, Miller JS et al. Evaluation of KIR ligand incompatibility in mismatched unrelated donor hematopoietic transplants. Killer immunoglobulin-like receptor. Blood 2002; 100: 3825–3827. | Article | PubMed | ISI | ChemPort |
35. Giebel S, Locatelli F, Lamparelli T, Velardi A, Davies S, Frumento G et al. Survival advantage with KIR ligand incompatibility in hematopoietic stem cell transplantation from unrelated donors. Blood 2003; 102: 814–819. | Article | PubMed | ISI | ChemPort |
36. Lowe EJ, Turner V, Handgretinger R, Horwitz EM, Benaim E, Hale GA et al. T-cell alloreactivity dominates natural killer cell alloreactivity in minimally T-cell-depleted HLA-non-identical paediatric bone marrow transplantation. Br J Haematol 2003; 123: 323–326. | Article | PubMed |
37. Bornhauser M, Schwerdtfeger R, Martin H, Frank KH, Theuser C, Ehninger G. Role of KIR ligand incompatibility in hematopoietic stem cell transplantation using unrelated donors. Blood 2004; 103: 2860–2861. | Article | PubMed | ISI |
38. de Santis D, Bishara A, Witt CS, Nagler A, Brautbar C, Slavin S et al. Natural killer cell HLA-C epitopes and killer cell immunoglobulin-like receptors both influence outcome of mismatched unrelated donor bone marrow transplants. Tissue Antigens 2005; 65: 519–528. | Article | PubMed | ChemPort |
39. Beelen DW, Ottinger HD, Ferencik S, Elmaagacli AH, Peceny R, Trenschel R et al. Genotypic inhibitory killer immunoglobulin-like receptor ligand incompatibility enhances the long-term antileukemic effect of unmodified allogeneic hematopoietic stem cell transplantation in patients with myeloid leukemias. Blood 2005; 105: 2594–2600. | Article | PubMed | ISI | ChemPort |
40. Ruggeri L, Capanni M, Mancusi A, Martelli MF, Velardi A. The impact of donor natural killer cell alloreactivity on allogeneic hematopoietic transplantation. Transplant Immunol 2005; 14: 203–206. | Article | ChemPort |
41. Davies SM, Kollman C, Anasetti C, Antin JH, Gajewski J, Casper JT et al. Engraftment and survival after unrelated-donor bone marrow transplantation: a report from the National Marrow Donor Program. Blood 2000; 96: 4096–4102. | PubMed | ISI | ChemPort |
42. Castro-Malaspina H, Harris RE, Gajewski J, Ramsay N, Collins R, Dharan B et al. Unrelated donor marrow transplantation for myelodysplastic syndromes: outcome analysis in 510 transplants facilitated by the National Marrow Donor Program. Blood 2002; 99: 1943–1951. | Article | PubMed | ISI | ChemPort |
43. Boeckh M, Nichols WG. The impact of cytomegalovirus serostatus of donor and recipient before hematopoietic stem cell transplantation in the era of antiviral prophylaxis and preemptive therapy. Blood 2004; 103: 2003–2008. | Article | PubMed | ISI | ChemPort |
44. Randolph SS, Gooley TA, Warren EH, Appelbaum FR, Riddell SR. Female donors contribute to a selective graft-versus-leukemia effect in male recipients of HLA-matched, related hematopoietic stem cell transplants. Blood 2004; 103: 347–352. | Article | PubMed | ISI | ChemPort |
45. Gordon RD, Simpson E, Samelson LE. In vitro cell-mediated immune responses to the male specific(H-Y) antigen in mice. J Exp Med 1975; 142: 1108–1120. | Article | PubMed | ISI | ChemPort |
46. den Haan JM, Sherman NE, Blokland E, Huczko E, Koning F, Drijfhout JW et al. Identification of a graft versus host disease-associated human minor histocompatibility antigen. Science 1995; 268: 1476–1480. | Article | PubMed | ChemPort |
47. Goulmy E, Schipper R, Pool J, Blokland E, Falkenburg JH, Vossen J et al. Mismatches of minor histocompatibility antigens between HLA-identical donors and recipients and the development of graft-versus-host disease after bone marrow transplantation. N Engl J Med 1996; 334: 281–285. | Article | PubMed | ISI | ChemPort |
48. Dickinson AM, Wang XN, Sviland L, Vyth-Dreese FA, Jackson GH, Schumacher TN et al. In situ dissection of the graft-versus-host activities of cytotoxic T cells specific for minor histocompatibility antigens. Nat Med 2002; 8: 410–414. | Article | PubMed | ISI | ChemPort |

Acknowledgements

We thank the technologists of the HLA typing laboratory and the laboratory for cellular histocompatibility testing of the Leiden University Medical Center.